Frustratingly Simplified Deployment in WLAN Localization by Learning from Route Annotation

Ryoma Kawajiri, Masamichi Shimosaka, Rui Fukui, Tonomasa Sato
Proceedings of the Asian Conference on Machine Learning, PMLR 25:191-204, 2012.

Abstract

Recently wireless LAN (WLAN) localization systems are gaining popularity in pervasive computing, machine learning and sensor networks communities, especially indoor scenarios where GPS coverage is limited. To accurately predict location, a large amount of fingerprints composed of received signal strength values is necessary. Moreover, standard supervised or semi-supervised approaches also require location information to each fingerprint, where annotation work is rather tedious and time consuming. To reduce the efforts and time required to build calibration data, we present a novel calibration methodology "route-annotation” and a self-training algorithm for learning from route information effectively. On the proposed calibration methodology, an annotator walks around while measuring fingerprints, then occasionally stops to annotate fingerprints with route from previous location to current location. This calibration reduces work time even compared to partially annotation, while routes have richer information for learning. The proposed learning algorithm comprises following two iterative steps: 1) inferring locations of each fingerprint under route constraints and 2) updating parameters. Experimental results on real-world datasets demonstrate learning from route-annotated data is comparable to state-of-the-art supervised and semi-supervised approaches trained with large amount of calibration data.

Cite this Paper

BibTeX
@InProceedings{pmlr-v25-kawajiri12, title = {Frustratingly Simplified Deployment in {WLAN} Localization by Learning from Route Annotation}, author = {Kawajiri, Ryoma and Shimosaka, Masamichi and Fukui, Rui and Sato, Tonomasa}, booktitle = {Proceedings of the Asian Conference on Machine Learning}, pages = {191--204}, year = {2012}, editor = {Hoi, Steven C. H. and Buntine, Wray}, volume = {25}, series = {Proceedings of Machine Learning Research}, address = {Singapore Management University, Singapore}, month = {04--06 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v25/kawajiri12/kawajiri12.pdf}, url = {https://proceedings.mlr.press/v25/kawajiri12.html}, abstract = {Recently wireless LAN (WLAN) localization systems are gaining popularity in pervasive computing, machine learning and sensor networks communities, especially indoor scenarios where GPS coverage is limited. To accurately predict location, a large amount of fingerprints composed of received signal strength values is necessary. Moreover, standard supervised or semi-supervised approaches also require location information to each fingerprint, where annotation work is rather tedious and time consuming. To reduce the efforts and time required to build calibration data, we present a novel calibration methodology "route-annotation” and a self-training algorithm for learning from route information effectively. On the proposed calibration methodology, an annotator walks around while measuring fingerprints, then occasionally stops to annotate fingerprints with route from previous location to current location. This calibration reduces work time even compared to partially annotation, while routes have richer information for learning. The proposed learning algorithm comprises following two iterative steps: 1) inferring locations of each fingerprint under route constraints and 2) updating parameters. Experimental results on real-world datasets demonstrate learning from route-annotated data is comparable to state-of-the-art supervised and semi-supervised approaches trained with large amount of calibration data.} }
Endnote
%0 Conference Paper %T Frustratingly Simplified Deployment in WLAN Localization by Learning from Route Annotation %A Ryoma Kawajiri %A Masamichi Shimosaka %A Rui Fukui %A Tonomasa Sato %B Proceedings of the Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2012 %E Steven C. H. Hoi %E Wray Buntine %F pmlr-v25-kawajiri12 %I PMLR %P 191--204 %U https://proceedings.mlr.press/v25/kawajiri12.html %V 25 %X Recently wireless LAN (WLAN) localization systems are gaining popularity in pervasive computing, machine learning and sensor networks communities, especially indoor scenarios where GPS coverage is limited. To accurately predict location, a large amount of fingerprints composed of received signal strength values is necessary. Moreover, standard supervised or semi-supervised approaches also require location information to each fingerprint, where annotation work is rather tedious and time consuming. To reduce the efforts and time required to build calibration data, we present a novel calibration methodology "route-annotation” and a self-training algorithm for learning from route information effectively. On the proposed calibration methodology, an annotator walks around while measuring fingerprints, then occasionally stops to annotate fingerprints with route from previous location to current location. This calibration reduces work time even compared to partially annotation, while routes have richer information for learning. The proposed learning algorithm comprises following two iterative steps: 1) inferring locations of each fingerprint under route constraints and 2) updating parameters. Experimental results on real-world datasets demonstrate learning from route-annotated data is comparable to state-of-the-art supervised and semi-supervised approaches trained with large amount of calibration data.
RIS
TY - CPAPER TI - Frustratingly Simplified Deployment in WLAN Localization by Learning from Route Annotation AU - Ryoma Kawajiri AU - Masamichi Shimosaka AU - Rui Fukui AU - Tonomasa Sato BT - Proceedings of the Asian Conference on Machine Learning DA - 2012/11/17 ED - Steven C. H. Hoi ED - Wray Buntine ID - pmlr-v25-kawajiri12 PB - PMLR DP - Proceedings of Machine Learning Research VL - 25 SP - 191 EP - 204 L1 - http://proceedings.mlr.press/v25/kawajiri12/kawajiri12.pdf UR - https://proceedings.mlr.press/v25/kawajiri12.html AB - Recently wireless LAN (WLAN) localization systems are gaining popularity in pervasive computing, machine learning and sensor networks communities, especially indoor scenarios where GPS coverage is limited. To accurately predict location, a large amount of fingerprints composed of received signal strength values is necessary. Moreover, standard supervised or semi-supervised approaches also require location information to each fingerprint, where annotation work is rather tedious and time consuming. To reduce the efforts and time required to build calibration data, we present a novel calibration methodology "route-annotation” and a self-training algorithm for learning from route information effectively. On the proposed calibration methodology, an annotator walks around while measuring fingerprints, then occasionally stops to annotate fingerprints with route from previous location to current location. This calibration reduces work time even compared to partially annotation, while routes have richer information for learning. The proposed learning algorithm comprises following two iterative steps: 1) inferring locations of each fingerprint under route constraints and 2) updating parameters. Experimental results on real-world datasets demonstrate learning from route-annotated data is comparable to state-of-the-art supervised and semi-supervised approaches trained with large amount of calibration data. ER -
APA
Kawajiri, R., Shimosaka, M., Fukui, R. & Sato, T.. (2012). Frustratingly Simplified Deployment in WLAN Localization by Learning from Route Annotation. Proceedings of the Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 25:191-204 Available from https://proceedings.mlr.press/v25/kawajiri12.html.

Related Material